-
Webserver - Google Cloud
-
Tunnel to expose local server to the internet - GCloud
ColorSpy is an app that lets you know the exact color your phone is seeing taking into consideration the different hues not just the plain of ROYGBV(Red Orange Yellow Green Blue Violet)
It leverages the power of data science. It takes an image from the front end, finds the most dominant color in the frame of view, allowing the user to move their device as much they want with no need to zoom in.
What it does
Detect with high accuracy the specific hue of color in the user's current frame of view
How we built it
- We used Unity to build the user interface and rendering
- For the Backend, we used:
- Flask: web server to handle API requests
- NumPy & Matplotlib: data analytics and color detection. it uses the Naive k-means algorithm to find the most dominant color in a given frame view.
Challenges we ran into
- Typecasting in python to properly serialize data from the front-end
- Lack of compatible open source libraries for encryption. RSA in unity has 512 bits minimum keysize and our python library had a 256 max keysize
Accomplishments that we're proud of
- Color detection with high accuracy
- Different hues
- Integration a data science into the project and learning more about it
- Teamwork
What we learned
“Anything that can go wrong, will go wrong”
- Keeping a network secure is hard and requires care
- Data serialization
- RSA and AES encryption
- Working in teams can be both a challenge and advantage
Built With
- google-cloud
- matplotlib
- numpy
- python
- unity
Log in or sign up for Devpost to join the conversation.